scholarly journals Use of Paraconsistent Feature Engineering to support the Long Term Feature choice for Speaker Verification

2021 ◽  
Vol 34 (1) ◽  
Author(s):  
Alex Marino Gonçalves De Almeida ◽  
Claudineia Helena Recco ◽  
Rodrigo Capobianco Guido
Keyword(s):  
Speech Synthesis ◽  
Speaker Verification ◽  
Wavelet Packet ◽  
Voice Conversion ◽  
Feature Engineering ◽  
Discrete Wavelet ◽  
Equal Error Rate ◽  
Discrete Wavelet Packet Transform ◽  
Open Research

The state-of-art models for speech synthesis and voice conversion can generate synthetic speech perceptually indistinguishable from human speech, and speaker verification is crucial to prevent breaches. The building feature that best distinguishes genuine speech between spoof attacks is an open research subject. We used the baseline ASVSpoof2017, Transfer Learning (TL) set, and Symlet and Daubechies Discrete Wavelet Packet Transform (DWPT) for this investigation. To qualitatively assess the features, we used Paraconsistent Feature Engineering (PFE). Our experiments pointed out that for the use of more robust classifiers, the best choice would be the AlexNet method, while in terms of classification regarding the Equal Error Rate metric, the best suggestion would be Daubechies filter support 21. Finally, our findings indicate that Symlet filter support 17 as the most promising feature, which is evidence that PFE is a useful tool and contributes to feature selection.

FUSION OF SURVEILLANCE IMAGES IN INFRARED AND VISIBLE BAND USING CURVELET, WAVELET AND WAVELET PACKET TRANSFORM

2010 ◽  
Vol 08 (02) ◽  
pp. 271-292 ◽  
Author(s):  
PARUL SHAH ◽  
S. N. MERCHANT ◽  
U. B. DESAI
Keyword(s):  
Wavelet Packet ◽  
Curvelet Transform ◽  
Visual Quality ◽  
Wavelet Packet Transform ◽  
Discrete Wavelet ◽  
Discrete Wavelet Packet Transform ◽  
Fused Image ◽  
Fusion Methods ◽  
Visible Images

This paper presents two methods for fusion of infrared (IR) and visible surveillance images. The first method combines Curvelet Transform (CT) with Discrete Wavelet Transform (DWT). As wavelets do not represent long edges well while curvelets are challenged with small features, our objective is to combine both to achieve better performance. The second approach uses Discrete Wavelet Packet Transform (DWPT), which provides multiresolution in high frequency band as well and hence helps in handling edges better. The performance of the proposed methods have been extensively tested for a number of multimodal surveillance images and compared with various existing transform domain fusion methods. Experimental results show that evaluation based on entropy, gradient, contrast etc., the criteria normally used, are not enough, as in some cases, these criteria are not consistent with the visual quality. It also demonstrates that the Petrovic and Xydeas image fusion metric is a more appropriate criterion for fusion of IR and visible images, as in all the tested fused images, visual quality agrees with the Petrovic and Xydeas metric evaluation. The analysis shows that there is significant increase in the quality of fused image, both visually and quantitatively. The major achievement of the proposed fusion methods is its reduced artifacts, one of the most desired feature for fusion used in surveillance applications.

Discrete Wavelet Packet Transform Based Discriminant Analysis for Whole Genome Sequences

2019 ◽  
Vol 18 (2) ◽  
Author(s):  
Hsin-Hsiung Huang ◽  
Senthil Balaji Girimurugan
Keyword(s):  
Discriminant Analysis ◽  
Wavelet Packet ◽  
Wavelet Packet Transform ◽  
Discrete Wavelet ◽  
Whole Genome ◽  
Statistical Classification ◽  
Genome Sequences ◽  
Discrete Wavelet Packet Transform ◽  
Alignment Free ◽  
Free Representation

Abstract In recent years, alignment-free methods have been widely applied in comparing genome sequences, as these methods compute efficiently and provide desirable phylogenetic analysis results. These methods have been successfully combined with hierarchical clustering methods for finding phylogenetic trees. However, it may not be suitable to apply these alignment-free methods directly to existing statistical classification methods, because an appropriate statistical classification theory for integrating with the alignment-free representation methods is still lacking. In this article, we propose a discriminant analysis method which uses the discrete wavelet packet transform to classify whole genome sequences. The proposed alignment-free representation statistics of features follow a joint normal distribution asymptotically. The data analysis results indicate that the proposed method provides satisfactory classification results in real time.

scholarly journals WAVELET ANALYSIS CONSIDERATIONS FOR EXPERIMENTAL NONSTATIONARY FLOW PHENOMENA

2016 ◽  
Vol 15 (1) ◽  
pp. 67
Author(s):  
M. L. S. Indrusiak ◽  
A. J. Kozakevicius ◽  
S. V. Möller
Keyword(s):  
Turbulent Flows ◽  
Wavelet Transforms ◽  
Fourier Spectrum ◽  
Transient Flow ◽  
Wavelet Packet ◽  
Discrete Wavelet ◽  
Velocity Measurements ◽  
Discrete Wavelet Packet Transform ◽  
Flow Phenomena ◽  
The Fourier Transform

In this work, wavelet transforms are the analysis tools for studying transient and discontinuous phenomena associated to turbulent flows. The application in quest results from velocity measurements with hot wire anemometry in the transient wake considering a circular cylinder in an aerodynamic channel. Continuous and discrete wavelet transforms are applied and compared with the corresponding results given by the Fourier transform. For the continuous wavelet transform, the Morlet function was adopted as transform basis, and for the discrete case, the Daubechies orthonormal wavelet with 20 null moments. Results using the discrete wavelet packet transform are also presented and compared. A wake past a cylinder was analytically simulated and compared with the actual one, both in transient flow. The ability of the wavelet transforms in the analysis of unsteady phenomena and the potential of the wavelet approach as a complementary tool to the Fourier spectrum for the analysis of stationary phenomena is presented and discussed.

Sign in / Sign up

Export Citation Format

Share Document